11© Copyright by International OCSCO World Press. All rights reserved. 2015 Research paper 11
of Achievements in Materialsand Manufacturing Engineering
International Scientific Journal
published monthly by the
World Academy of Materials
and Manufacturing Engineering
Volume 68 • Issue 1 • January 2015
Etching techniques for duplex steels
P. Martínek*, P. Podaný COMTES FHT a.s., Průmyslová 995, 334 41 Dobřany, Czech Republic
* Corresponding e-mail address: [email protected]
ABSTRACT
Purpose: This contribution describes a search for an optimum etching technique for duplex ferritic-austenitic steels which would enable metallographers to find the fractions of major phases by image analysis and determine the amounts, distributions and types of intermetallic phases.
Design/methodology/approach: The microstructures were revealed by etching with seven different reagents. The phase compositions were evaluated using either image analysis or test grid-based quantitative analysis. The experimental materials were X2CrNiMoN 22-5-3 steel and its cerium-doped variant. Each of them was examined in two conditions: upon open-die forging and subsequent solution annealing and upon long-time annealing, where the latter led to extensive precipitation of intermetallic phases. The fractions of major and intermetallic phases were also determined using EBSD. Both quantitative and qualitative EBSD data were then compared to the values obtained using optical microscopy.
Findings: In X2CrNiMoN 22-5-3 duplex steel, the microstructure can be revealed using various reagents and both chemical and electrochemical etching. The differences between the reagents, when used for evaluating the amounts of major phases (austenite + ferrite), were not substantial. The fraction of sigma phase in long-time-annealed samples can be evaluated using image analysis only if etched with NaOH solution or NH4OH. These etchants also effectively reveal carbides on grain boundaries. However, the values obtained with NaOH are overestimated. When the other reagents are used, the evaluation must be done using another method (e.g. grid-based quantitative analysis). Sigma phase proportions found by optical microscopy are higher than those measured using EBSD. In order to identify microstructural variations across the forged parts, specimens were taken from three locations (centre, ¼ , edge). The sigma phase amounts found in all three testing locations of the sample of the cerium-doped duplex steel were higher than the corresponding amounts in the cerium-free sample. In both materials, the amounts of sigma phase are higher in the centre of the sample than near the edge. This difference is more significant in the cerium-free material.
Originality/value: The article is devoted to phase content evaluation in duplex steels by different methods. EBSD and image analysis of micrographs were compared. Micrographs were acquired by light microscopy after microstructure revelation by various etchants. The content of ferrite, austenite and intermetallic phases was evaluated.
Keywords: Duplex steels; Etching techniques
Reference to this paper should be given in the following way:
P. Martínek, P. Podaný, Etching techniques for duplex steels, Journal of Achievements in Materials and Manufacturing Engineering 68/1 (2015) 11-16.
MATERIALS
Research paper12
Journal of Achievements in Materials and Manufacturing Engineering
P. Martínek, P. Podaný
1. Introduction
The history of duplex steels, i.e. two-phase austenitic-
ferritic stainless steels, is almost as long as that of stainless
steels but the interest of the industry in this group of steels
has been increasing recently. This mainly holds for
applications where austenitic steels do not provide
a guarantee of fault-free and safe operation, particularly in
environments where stress-corrosion cracking may occur
[1]. Duplex stainless steels are frequently used in industry
for their excellent combination of mechanical properties
and corrosion resistance. Their resistance to uniform
corrosion is similar to austenitic steels but their strength is
much higher [2]. These characteristics depend on the two-
phase microstructure that comprises austenite and delta
ferrite [1]. Ferrite enhances strength and imparts resistance
to stress-corrosion cracking. However, ferrite is also prone
to microstructural changes whereas austenite remains the
stable phase. As a result, these steels develop undesirable
intermetallic phases in the critical temperature range of
650-900°C during forging or rolling earlier than austenitic
steels [3, 4]. This applies primarily to the sigma phase (�)
which is hard, brittle, non-magnetic and stable.
Precipitation of sigma phase in steel increases the
brittleness, hardness, ultimate and yield strengths whereas
the elongation and reduction of area in ambient-
temperature tensile test decrease.
In order to make products that provide trouble-free
operation, one needs to have detailed knowledge of the
microstructure of duplex steels. The proportions of major
phases must be known, as well as their amounts, sizes,
distributions and also the types of intermetallic phases, if
present. To facilitate mapping of these characteristics, the
present experiment was proposed. Its aim was to identify
the most appropriate etching technique for these steels to
meet the aforementioned need.
2. Experimental
The samples used for examination were forged bars
whose microstructures were in two conditions. One of them
was an open-die forged and solution-annealed condition.
The forging operations were performed between 1200°C
and 900°C. The soaking time at the forging temperature
was 10 hours. The solution annealing was carried out at
1050°C for 4 hours with subsequent quenching in water.
The other condition of the samples was achieved by the
same procedure with additional annealing at 750°C for
48 hours and subsequent cooling in air. The goal was to
induce extensive precipitation of intermetallic phases. In
the forged and solution-annealed specimens, the fractions
of austenite and ferrite were evaluated. In long-time-
annealed specimens, the evaluation focused primarily
on intermetallic phases. With the aim of identifying
microstructural variations across the forged parts,
specimens were taken from three locations (centre, ¼,
edge) - as shown in Fig. 1.
Fig. 1. Specimen codes
The experimental materials were X2CrNiMoN 22-5-3
steel and its cerium-doped variant (the specimen code
included the letters Ce).
2.1. Optical microscopy
The microstructures of the duplex steel specimens were
revealed using several reagents which fall into two groups.
The first one comprises chemical reagents: Beraha II+
K2S2O5, Beraha II and Murakami's reagent. The second
includes electrolytic etchants: NaOH, NH4OH, 60% HNO3
and oxalic acid. Once the microstructure was revealed, the
fractions of individual microstructure constituents were
evaluated. The number of fields of view used for the
evaluation was 20 in both groups of specimens and for all
etchants.
An effort was made to employ NIS Elements 3.2 image
analysis software for the evaluation in all cases.
In solution-annealed specimens, the phase composition
was evaluated by image analysis upon etching with the
following reagents: Beraha II+K2S2O5, NaOH and
Murakami’s reagent. The other etchants used (oxalic acid,
60% HNO3 and Beraha II) proved unsuitable for
quantitative image analysis evaluation, as they did not
provide sufficient contrast between phases.
In the long-time-annealed specimens, quantitative
image analysis could only be used for evaluating the phase
composition upon etching with NaOH solution or NH4OH
or Murakami’s reagent. When the other etchants (Table 1)
were used, grid-based quantitative analysis had to be
employed. In the grid-based analysis, 20 fields of view
were used as well.
1. Introduction
2. Experimental
2.1. Optical microscopy
13Etching techniques for duplex steels
Volume 68 • Issue 1 • January 2015
Table 1.
Sigma phase fractions in long-time-annealed specimens [%]
Specimen
Etching reagent
Beraha II +
K2S2O5
NaOH Beraha II
�+chi phase
fraction
�+chi phase
fraction
�+chi phase
fraction
A 18.94 ± 1.56 20.25 ± 1.39 17.50 ± 1.16
B 20.58 ± 2.87 18.99 ± 1.90 not evaluated
C 13.84 ± 1.43 15.75 ± 1.28 14.67 ± 1.28
Ce A 22.05 ± 2.04 23.50 ± 1.53 21.16 ± 1.43
Ce B 19.57 ± 1.77 20.18 ± 1.54 not evaluated
Ce C 21.26 ± 1.67 19.69 ± 1.20 21.06 ± 1.39
Specimen
Etching reagent
Murakami NH4OH
�+chi phase fraction �+chi phase fraction
A not evaluated 20.03 ± 1.04
B 18.08 ± 1.42 0.94 ± 0.94
C 12.80 ± 1.40 1.00 ± 1.00
Ce A not evaluated 21.86 ± 1.46
Ce B 21.24 ± 1.66 1.36 ± 1.36
Ce C 19.19 ± 1.36 1.24 ± 1.24
2.2. EBSD analysis
Crystallographic data on the specimens were gathered
using EBSD (Electron Backscatter Diffraction). This
method relies on analysing Kikuchi lines obtained by
directing an electron beam on a specimen tilted under a
large angle in the chamber of a scanning electron
microscope (SEM). Using the EBSD method, the fractions
of phases were determined. The values were then compared
to the data obtained using optical microscopy. The scope of
measurement in each method was different. Under
the optical microscope, the phase fractions were
determined from 20 fields of view at a magnification of
500. For the EBSD analysis, five fields of view were used,
each with the size of 400×200 µm. By area, one field of
view in EBSD analysis approximately corresponds to
3.5 fields of view under the optical microscope at the 500×
magnification. The aggregate areas evaluated by optical
microscopy and EBSD are thus roughly comparable. An
example EBSD map used for evaluating the sigma phase
fraction is shown in Fig 2. The phase composition found
by EBSD is listed in Table 2.
Fig. 2. Specimen C - upon long-time annealing. EBSD
analysis; ferrite appears red. � phase appears yellow.
Austenite - colours depending on orientation
Table 2.
Phase fractions found by EBSD in long-time-annealed
specimens [%]
Specimen Ferrite Austenite �+chi
A 12.5 72.8 14.7
B 9.2 75.6 15.2
C 17.7 71.2 11.1
Ce A 7.5 68.3 24.2
Ce B 21.0 68.1 19.9
Ce C 5.8 75.1 19.1
3.Discussion
3.1. Solution-annealed specimens
Duplex steels can be etched both chemically and
electrolytically using several different etchants. In the
solution-annealed specimens, no intermetallic phases were
found using either optical or scanning electron microscopy.
Therefore, only the ferrite and austenite fractions were
evaluated for the solution-annealed specimens.
The differences between phase fractions found after
etching with different reagents are small and close to the
measurement error. The ferrite amount is between 43 and
46%. Table 3 shows that the amount of ferrite found upon
etching with NaOH was higher than with the other two
reagents. It can be explained by the fact that etching with
NaOH produces a relatively thick oxide layer on ferrite
grains. This layer thus overhangs other phases and distorts
the data.
The Beraha II+K2S2O5 reagent is a colour etchant which
produces a thin oxide layer on the specimen surface. This
3. Discussion
2.2. EBSD analysis
3.1. Solution-annealed specimens
Research paper14
Journal of Achievements in Materials and Manufacturing Engineering
P. Martínek, P. Podaný
layer is usually non-uniform. Additional etching is
impossible with this reagent. Etching with Murakami’s
solution is carried out at an elevated temperature and the
etching time is longer than with other reagents (Fig 3).
Etching with NaOH does not involve such disadvantages -
Fig. 4. It is an electrolytic process which was found to be the
most feasible one. On the other hand, it may lead to a slight
distortion of results.
The average amounts of ferrite across all reagents and
measurement areas in cerium-free and cerium-doped
specimens were 42.6% and 45.6%, respectively. The
difference between the two materials is approximately three
percentage points. This difference is probably the
consequence of the difference in their chemical
compositions. No differences were found between locations
from which specimens were taken.
Table 3.
Ferrite amounts in solution-annealed samples [%]
Specimen
Etching reagent
Beraha II + K2S2O5 NaOH Murakami
Ferrite fraction Ferrite fractionFerrite fraction
A 41.45 ± 0.73 43.65 ± 0.68 42.70 ± 1.07
B 41.32 ± 0.75 43.80 ± 0.80 41.09 ± 0.70
C 43.88 ± 0.46 43.87 ± 0.69 41.94 ± 0.66
Ce A 47.02 ± 0.80 46.32 ± 1.00 45.66 ± 0.98
Ce B 43.96 ± 0.67 45.44 ± 0.69 45.97 ± 0.66
Ce C 44.63 ± 0.68 45.60 ± 0.63 46.12 ± 0.73
Fig. 3. Specimen CeB - upon solution-annealing. Etched
with Murakami’s reagent
Fig. 4. Specimen A – upon solution-annealing. Etched with
NaOH solution
3.2. Long-time-annealed specimens
Long-time annealing led to extensive precipitation
of intermetallic phases. Carbides were not evaluated using
optical microscopy and image analysis. Although they were
readily visible upon etching with some reagents, they were
too small for quantitative evaluation. This is why the
amounts of sigma phase and chi phase were the main
variables evaluated for various reagents. Both phases are
very similar in terms of their properties. They are impossible
to distinguish from each other using optical microscopy.
Another reason why both phases were evaluated as a single
one and denoted in the tables below as “sigma phase” is that
their effects on the properties of duplex steels are similar.
The etch with Beraha II+K2S2O5 or Beraha II revealed
both major phases and sigma phase relatively well.
However, carbides and austenite grain boundaries were not
visible.
When NaOH, NH4OH and Murakami’s reagent were
used, particles of sigma phase became prominent and well-
suited for image analysis; and carbides on grain boundaries
were visible (Figs. 5-8). However, as Table 1 suggests, the
highest fraction of sigma phase was found upon the NaOH
etch (Fig. 6). Etching with this solution produces a relatively
thick oxide layer on the surface. This oxide layer causes the
measured values for a particular phase to be higher than the
actual values. The advantage of this reagent lies in that the
sigma phase proportion can be effectively measured by mean
of image analysis.
Upon etching with nitric acid or oxalic acid, sigma
phase could only be evaluated using the grid-based analysis
Ferrite
Austenite
Austenite
Ferrite
3.2. Long-time-annealed specimens
15Etching techniques for duplex steels
Volume 68 • Issue 1 • January 2015
but not by image analysis. Austenite grains and carbides
on their boundaries were well visible.
The sigma phase amounts found in all three testing
locations of the sample of the cerium-doped duplex steel
were higher than the corresponding amounts in the cerium-
free sample. This is probably due to the higher ferrite
amount in this material, as sigma phase nucleates more
readily in ferrite than in austenite.
In both materials, the amounts of sigma phase were
higher in the centre of the sample than near the edge. This
difference was more significant in the cerium-free material.
Fig. 5. Specimen A - upon long-time annealing. Etched
with Beraha II+K2S2O5
Fig. 6. Specimen A - upon long-time annealing. Electrolytic
NaOH etch
Fig. 6. Specimen C - upon long-time annealing. Etched with
Murakami’s reagent
Fig. 8. Specimen CeC - upon long-time annealing. Etched
with Beraha II
3.3. Comparison of results obtained using
optical microscopy and EBSD
The comparison of data shown in Tables 2 and 3
revealed that EBSD analysis yielded lower sigma phase
values than image analysis. The possible causes are given
below.
When the amount of a certain phase is evaluated using
optical microscopy and image analysis, the value tends to
be overestimated. This is due to both human factor and
etching products. Those form on the surface of the phase
Austenite
Ferrite
� phase
Austenite
Ferrite
� phase
Austenite
Ferrite
Austenite
Ferrite
� phase
3.3. Comparison of results obtained using opti-cal microscopy and EBSD
Research paper16 READING DIRECT: www.journalamme.org
Journal of Achievements in Materials and Manufacturing Engineering
by chemical or electrochemical reactions. They may have
relatively large thickness and may extend beyond the phase
in question, thus distorting the measurement.
On the other hand, the EBSD outcomes depend on the
quality of diffraction patterns of individual phases. As the
sigma phase particles are small and the diffraction is less
clear on the interphase interfaces, the sigma phase is not
identified on the entire area. As a result, its area fraction
values are underestimated.
The eutectoid decomposition of ferrite produces sigma
phase and small austenite grains which are difficult to
discern under the optical microscope. Consequently, they
tend to be included in the sigma phase area. EBSD
analysis, however, allows even these small grains to be
identified.
4. Conclusions
The purpose of this study was to find the optimum
etching technique for duplex steels which would enable a
metallographer to determine the amounts of delta ferrite
and austenite and the fractions, distributions and types of
intermetallic phases using image analysis. The values
obtained by means of optical microscopy were compared to
the EBSD data. The results can be summarised as follows.
In X2CrNiMoN 22-5-3 duplex steel, the microstructure
can be revealed using various reagents and both chemical
and electrochemical etching.
The differences between the reagents when used
for evaluating the amounts of major phases (austenite
+ferrite) were not substantial.
The fraction of sigma phase in long-time-annealed
samples can be evaluated using image analysis only if etched
with NaOH solution or NH4OH. These etchants also
effectively reveal carbides on grain boundaries. However,
the values obtained with NaOH are overestimated. When the
other reagents are used, the evaluation must be done using
another method (e.g. grid-based quantitative analysis).
Sigma phase proportions found by optical microscopy
are higher than those measured using EBSD.
In order to identify microstructural variations across
the forged parts, specimens were taken from three locations
(centre, ¼, edge). The sigma phase amounts found in all
three testing locations of the sample of the cerium-doped
duplex steel were higher than the corresponding amounts in
the cerium-free sample.
In both materials, the amounts of sigma phase are
higher in the centre of the sample than near the edge. This
difference is more significant in the cerium-free material.
Acknowledgements
This paper was prepared under the project entitled
Development of West-Bohemian Centre of Materials and
Metallurgy No.: LO1412, which is funded by the Ministry
of Education of the Czech Republic.
References
[1] M. Liljas, 80 years with duplex steels, a history review
and prospects for the future, http://www.euroinox.org
/pdf/map/paper/Duplex_Steels_EN.pdf
[2] Stainless Steel Conference Science and Market, ISBN
91-974131-9-4, Helsinki, Finland, 2008.
[3] Information on http://www.mmspektrum.com.
[4] T. Chen, K. Weng, J. Yang, The effect of high-
temperature exposure on the microstructural stability
and toughness property in a 2205 duplex stainless
steel Materials Science and Engineering, 338/1-2
(2002) 259-270.
[5] G. Kaishu, Effect of aging at 700°C on precipitation
and toughness of AISI 321 and AISI 347 austenitic
stainless steel welds, Nuclear Engineering and Design
235 (2005) 2485-2494.
[6] J.C. Dutra, F. Siciliano, A.F. Padilha, Interaction
between second-phase particle dissolution and
abnormal grain growth in an austenitic stainless steel,
Materials Research 5/3 (2002) 379-384.
References
Acknowledgements
4. Conclusions